Innovative Projects Realized

Oscillation-based Fuel Cell Diagnostics

In this project, we propose two diagnostic tools that can identify dynamical processes in various fuel cell operating regimes, using the difference in the time constant of these processes. For example, conductive transport of electrons is faster than diffusive transport of gasses. We oscillate current and pressure at different frequencies, and measure the cell voltage. We use the amplitude ratio and phase different of these oscillations to detect dynamical processes in the fuel cells. Specifically, we are interested to study water transport in the catalyst pores and hydrogen transfer leak through membrane pinholes using pressure and current oscillations, respectively. These diagnostic tools enable Greenlight to build fuel cell test stations with enhanced capabilities.

Super Resolution Model for License Plate Recognition

Our novel method enhances low resolution images from surveillance footage and facilitates automatic recognition of license plates. Even if the frames are blurry and unclear, the proposed model can enhance while prioritizing character and text information detection. This will be beneficial to security, law enforcement and investigation agencies.

Beyond Keywords: Semantic Search Framework for Data in Organizations

Next to its essential role of supporting operational and decisional business activities, data also has economic significance. The desire to seek maximum value from their data assets prompts organizations to implement different infrastructures, architectures, governance, and security to facilitate creating and storing huge volume and variety of data. However, these implementations come with challenges as the data is not often the main focus, and this usually makes data management and discovery difficult. The data mesh paradigm advocates for data to be decomposed around domains and served as a product for use by data users. Having data as products is posited to enable ease in data management and discovery. This also enables the application of semantic searches to data ecosystem. Integrating recent advances in artificial intelligence into semantic search mechanisms makes them great candidates to enable business to derive optimal value from their data assets. In this regard, our project’s goal is to develop a framework that utilizes ontology embeddings, vector search, and large language models (LLMs) to improve data discovery and management using semantic search. The result will be capabilities that enable organizations to manage their data products through more intelligent and customized searches of their data.

Automated Attack Hypothesis and Testflow Generation

In the field of cybersecurity, it is increasingly important to actively find and stop security threats, a process called threat hunting. The threat hunting is often performed manually, which is a tough process the requires deep knowledge, lots of experience, and time, which can lead to missed attacks and slow responses, affecting companies and countries. This research suggests using an automated system to make threat hunting faster and more efficient. The goal is to change how threat hunting is performed by using a system that automatically creates hypotheses and test plans. It will improve how quickly and accurately threats are found and dealt with, making cybersecurity stronger. The plan is to build a new system using advanced techniques to automatically suggest possible threats from network data and user activities, then create test plans to check these threats. This system will mix methods to understand complex data and use machine reasoning to think like humans. The aim is to create a smart system that adjusts to new cyber threats…

Development of a Strategy to Design and Optimize a Non-Licensed UHF Band System for Operations under Harsh Environment for Localization Applications

Two antennas will be designed at 915 MHz to be installed in a small cavity within the body of an equipment used for underground operations. The two antennas are expected to provide reasonable operating bandwidth and radiation efficiency to radiate through lossy polyurethane used for protection, metallic cavity surrounding the antennas, and the conductive equipment body. The polyurethane has a high dielectric constant which reduces the antenna size, and lossless material will also be used around the antennas to enhance its radiation efficiency. The antennas will be installed at optimal equipment positions where the radiation efficiency is sufficiently high. This project will produce a new product which can be used to localize an underground equipment as fast as possible to reduce costs and maintenance time. Also, a lot of companies will be interested in this low-cost compact product, hence, the partner’s marketing strategy will be supported and developed. Also, the research and development department of the partner organization will be improved by including the outcomes of the proposed project.

The state of literacy in Nova Scotia: an interactive dashboard

The project aims to develop an interactive dashboard to accompany a report on the state of literacy in Nova Scotia to be produced by Literacy Nova Scotia (LNS), as a means of quickly communicating key information for policy decision-makers and internal stakeholders. Through a participatory design program, we will engage with internal and external stakeholders to design a dashboard that will address the user’s information-seeking needs. In contrast with usability evaluation, we will focus on learning and comprehension as measures for evaluating the effectiveness of the dashboard. Expected outcomes will be a working dashboard for data exploration for LNS stakeholders.

Leveraging AI Techniques to Analyze Corporate Documentation for ESG Indicators

Significant demand exists by society for information about how companies are conducting themselves on matters of environmental stewardship, social responsibility, and good corporate governance (ESG). Typically, however, such information is hampered by significant inconsistencies in availability, which hinders access to reliable information. Our goal in this research project is to address these obstacles faced by business, civil society, and governmental stakeholders in making informed decisions on a given company or industry. Our project will use the power of information technology to develop a software tool designed to ethically collect, organize, augment, and evaluate ESG data for a vast number of companies and industries to give a more complete picture of company/industry ESG activities. In doing so, this project addresses a significant research gap across information science and business by advancing research that provides measurable benefit to information stakeholders reliant on this data to make effective decisions

The economic, environmental, and social implications of implementing smart grid management systems

Smart grid energy systems are considered the future of electrical grid management and delivery. These technologies and programs are innovative ways to ensure a reliable electricity grid, promote renewable energy, increase efficiency, and enhance consumer control. Smart grids operate as automated or computerized systems that centralize electricity grid control and ultimately provide a more resilient power grid. The research will examine how organizations, businesses, and jurisdictions can transition from a conventional grid management system to smart grid technologies. An in depth analysis will be conducted regarding government programs and initiatives to compensate parties who have invested in smart grid energy management systems. Additionally, social and environmental impacts will be assessed, on both a provincial and federal level, to determine the benefits of a transition to smart grid technology. Eco-Shift Power will gain essential knowledge of developments and market trends in an ever-evolving energy sector. The research will provide their business with relevant information on the smart grid approach, how to implement the technology, and available incentive programs. This will add tremendous value to their business model in terms of energy cost reduction, corporate social responsibility, and improving sustainability.

Evaluating the Association of Intermediate Social Determinants of Health, Dementia Modifiable Risk Factors, and Cognitive Health for On-Reserve First Nations Peoples

The incidence of dementia, a clinical syndrome marked by progressive cognitive decline, is rising more steadily among Indigenous populations in Canada, mainly due to modifiable risk factors, such as diabetes and physical inactivity. Previous research has found that cultural and social connection and access to healthcare impact the incidence of many of the modifiable risk factors of dementia among the general populations and global Indigenous peoples. However, little research has investigated the association between these factors, the modifiable risk factors of dementia, and cognitive health among Indigenous peoples. Our study investigates the relationship between connection to culture, social support, and access to healthcare, the modifiable risk factors of dementia, and cognitive health among on-reserve First Nations peoples in Canada. Results will inform future community programming, research to develop interventions and public policy.

Establishing the Molecular Weight and the Length of Grafted Chains of PDMS Brushes Using the Ellipsometry Method

The field of research exploring slippery covalently attached liquid surfaces (SCALS), formed with Polydimethylsiloxane (PDMS) brushes, is fairly new and is proving pivotal in creating economically feasible and non-toxic solutions to resist fouling and prevent ice adhesion. These solutions will contribute to a more sustainable future by lowering the energy expenditure required for optimal efficiency of systems such as solar panels, wind turbines and aircraft wings.

The research aims to develop a simple method to establish the molecular weight and length of grafted chains, through performing a series of in-situ ellipsometry measurements of PDMS brushes in different solvents. A series of values and swelling ratios will be compiled and compared with current polymer theory to characterize the molecular weight and length of the grafted chains of the PDMS brushes. The proposed methodology could prove to be faster, more accessible and potentially just as accurate as current methods, including AFM force measurements and neutron reflectometry. If it is found that using this expeditious process to understand the brushes’ properties is possible and accurate, designs can be created with a high degree of assurance and knowledge about how the SCALS will behave.

Integrated organic opto-bioelectronics sensors

We are now in Polytechnique Montreal studying organic electrochemical transistors (OECT), which can be used to detect the analyte ion concentration in the liquid environment. The nature of an OECT is to convert ion concentration signal to electron signal. In this short-term cooperation, we aim to integrate OECT with an organic light-emitting diode (OLED), so that the obtained electrical signal by OECT can be further converted to photon signal. This renders it possible to convert ion signal directly to photon signal. Since both OECT and OLED are flexible, printable and wearable, so, this project may pave the way to bioactive, flexible and printable organic light emission sensors.

Temporal and spatially consistent inpainting with deep diffusion models

Most current brain imaging software is designed to work with images that have a consistent anatomy. However, this can be a problem when there are changes in the brain, like the appearance or disappearance of lesions, between two scans taken at different times. While we could develop complex new software for these situations, many researchers prefer to use existing, well-known tools, especially for studies over time. One suggested solution is to use a technique called image inpainting. In image inpainting, we fill in certain parts of an image based on a special pattern, making these parts look like they naturally belong there. This is really useful in brain scans, for example, when we need to add in images of lesions from multiple sclerosis (MS) in a way that looks real and fits with the rest of the brain image. Current inpainting methods usually practice on specific patterns, which means they might not work well on new or different types of patterns. Plus, these methods often struggle with handling lesions of various sizes and locations because they’re used to working with the same types of patterns, which limits how well they can work on different or new ones.